Some basic questions as a new user of the package

106 views
Skip to first unread message

Soumya Banerjee

unread,
Oct 19, 2022, 9:15:29 AM10/19/22
to unma...@googlegroups.com
Dear all, 

I am intending to use unmarked for running single-season occupancy models on detection/nondetection data on striped hyenas collected in a human dominated landscape in western India. Having gone through Fiske & Chandler 2011 and the vignettes, I have two questions which are rather basic, so I apologize if these have already been covered by the above material: 

i) How do I generate heat maps of occupancy probabilities given a map of the study area? Of course, I could obtain the corresponding probabilities based on the best fitting model/model averaged values and the covariate values at each site and then map the data on R or a GIS platform, but I wanted to know if there is any functionality in unmarked that would make the process simpler. 

ii) Does unmarked accommodate the Hines et al. 2010 model, which deals with spatially autocorrelated detections across replicates? In my study, the replicates were contiguous segments of transects 200-300m long, and I anticipate that hyenas may have moved across contiguous segments, and thus I would like to use this model. 

I would be highly grateful if these queries could be addressed at the earliest. 

Best Regards, 

Soumya Banerjee, 
Junior Research Fellow, 
Wildlife Institute of India, 
Dehradun, Uttarakhand, India

Ken Kellner

unread,
Oct 19, 2022, 9:20:33 AM10/19/22
to unmarked
Hi Soumya,

i) Yes, you can supply map data (as a RasterStack) to the newdata argument of the predict() function to generate a map of occupancy probability from a fitted model. See the species distribution vignette (https://cran.r-project.org/web/packages/unmarked/vignettes/spp-dist.html) for more details.

ii) Unfortunately unmarked does not currently support the Hines et al. 2010 model, although we would like to in the future. I believe PRESENCE software does support this model.

Ken

Marc Kery

unread,
Oct 19, 2022, 9:48:17 AM10/19/22
to unma...@googlegroups.com
Dear Soumya,

there is no in-built functionality in unmarked to create such maps. So you will have to generate them as you suggest.

And unfortunately, the Hines et al. 2010 ("tiger on trails"-) model is not currently supported by unmarked. You can fit it with PRESENCE and presumably also with MARK. Or use JAGS and code it up yourself 🙂

Best regards  --- Marc


From: unma...@googlegroups.com <unma...@googlegroups.com> on behalf of Soumya Banerjee <banerjee...@gmail.com>
Sent: Wednesday, October 19, 2022 08:22
To: unma...@googlegroups.com <unma...@googlegroups.com>
Subject: [unmarked] Some basic questions as a new user of the package
 
--
*** Three hierarchical modeling email lists ***
(1) unmarked (this list): for questions specific to the R package unmarked
(2) SCR: for design and Bayesian or non-bayesian analysis of spatial capture-recapture
(3) HMecology: for everything else, especially material covered in the books by Royle & Dorazio (2008), Kéry & Schaub (2012), Kéry & Royle (2016, 2021) and Schaub & Kéry (2022)
---
You received this message because you are subscribed to the Google Groups "unmarked" group.
To unsubscribe from this group and stop receiving emails from it, send an email to unmarked+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/unmarked/CAHz3eBdriOS%3DTib8g3y2SFgYFFZD96vRe0M7x_W8QQzs9JNV4A%40mail.gmail.com.

Marc Kery

unread,
Oct 19, 2022, 9:52:03 AM10/19/22
to unma...@googlegroups.com
Ah, well, my first reply is partly wrong; please see Ken's earlier email.

Marc


From: unma...@googlegroups.com <unma...@googlegroups.com> on behalf of Marc Kery <marc...@vogelwarte.ch>
Sent: Wednesday, October 19, 2022 15:48
To: unma...@googlegroups.com <unma...@googlegroups.com>
Subject: Re: [unmarked] Some basic questions as a new user of the package
 

Soumya Banerjee

unread,
Oct 20, 2022, 1:17:58 AM10/20/22
to unmarked
Thank you very much :) 
Reply all
Reply to author
Forward
0 new messages